Nunberg maintains that there are cases like ''I am traditionally entitled to a last meal'', as uttered by a condemned prisoner facing the firing squad, which suggest that an indexical like ‘I’ does double duty as a vehicle of singular and general reference. I argue against this claim. My position is that the sentence should be factored out into two: ''Traditionally, a condemned prisoner is entitled to a last meal'' and ''I am a condemned prisoner''. Nunberg’s sentence is generated by means of an illicit substitution of ‘I’ for ‘a condemned prisoner’ inside the scope of ''traditionally''. The morale is that sloppy or literally nonsensical speech like Nunberg’s sentence is not suitable as data for logical analysis of natural language. What is suitable data is the two-premise argument I put forward., Nunberg tvrdí, že existují případy, jako například ,,tradičně mám nárok na poslední jídlo'', jak vyslovil odsouzený vězeň, který čelí střelbě, což naznačuje, že indexové označení ,,I'' má dvojí povinnost jako prostředek jednotného a obecného odkazu. Proti tomuto tvrzení argumentuji. Můj postoj je takový, že věta by měla být zohledněna ve dvou: ,,Odsouzený vězeň má tradičně nárok na poslední jídlo'' a ,,Jsem odsouzený vězeň''. Nunbergův rozsudek je vytvářen nedovolenou náhradou slova ,,I'' za ,,odsouzeného vězně'' v rozsahu ,,tradičně''. Morálka je, že nedbalý nebo doslova nesmyslný projev jako Nunbergova věta není vhodný jako data pro logickou analýzu přirozeného jazyka. Vhodnými daty jsou argumenty, které jsem navrhl., and Bjørn Jespersen
With the gradual improvement of the telecommunication infrastructure in China, the Internet and other new technologies have been frequently used. The Internet technology also brings many network security threats, for example, botnet, while bringing convenience. Botnet is a network formed between hosts controlled by malicious code. One of the most serious threat to network security faced by the Internet is a variety of malicious network attacks on the carrier of botnet. Back propagation (BP) neural network is proposed to detect botnet threat transmission. In this study, a botnet detection model was established using BP neural network system. BP neural network classifier could identify the botnet traffic and normal traffic. Moreover a test was carried out to detect botnet traffic using BP neural network; the performance of the BP neural network classifier was evaluated by the detection rate and false positive rate. The results showed that it had high detection rate and low false positive rate, which indicated that the BP neural network had a good performance in detecting the traffic of botnet threat transmission.